Skip to content

IncarnaMind

Chat with your own PDF and TXT documents through LLMs

Visit official site

What IncarnaMind is

The project is described as an open-source way to chat with personal documents through large language models, using a sliding window chunking scheme and an ensemble retriever.

What you can do with it

  • Ask questions across several PDF and TXT files at once
  • Run simple and multi-hop queries over a document set
  • Balance fine and coarse access with adaptive windows

Requirements and setup

  • The GGUF version needs more than 35GB of GPU RAM
  • Setup uses a Conda environment and installed dependencies
  • An API key is placed in configparser.ini

Pros & cons

✓ What we like

  • Multi-document chat removes the single-file limit
  • Local open models are supported alongside hosted ones

! What to watch out for

  • Running the GGUF version needs more than 35GB of GPU RAM
  • Setup asks for an API key in configparser.ini

FAQ

Which file formats are supported?

The material lists PDF and TXT files.

Which models can it use?

It names OpenAI GPT, Anthropic Claude, Llama2 and other open LLMs.

Last reviewed: 2026-11-15

More AI chatbot tools

View all →

How we review